How to I stop a runaway lpsched?

[Cindy Wolford]

Hi there,

I have noticed that I have a runaway lpsched. I'd like to kill it because it is using a lot of CPU, but I don't know what to kill.

When I do a ps -fu lp:
lp 767 1 228 Oct 1 ? 36110:55 /usr/sbin/lpsched
lp 23292 1 0 15:26:20 ?
0:06 lpsched

If I do a top command, I can see that it is taking almost 90% of a CPU.

How do I stop this? It appears to have been running for days....

TIA
Cindy

[Rita C Workman]

to stop the spooler

lpshut

You can restart it with

lpsched

Rita

[James R. Ferguson]

Hi Cindy:

The proper way to stop it is 'lpshut'.

Regards!

...JRF...

[Uday_S_Ankolekar]

Hi,

Try lpshut to stop and restart it by lpsched.
This would be best way kill your runaway process

Good Luck..
-USA..

[Cindy Wolford]

Hi,

I should have mentioned that I had tried stopping and restarting the spooler....but this did not resolve my problem.

Any other thoughts, I'd appreciate it.

Cindy

[A. Clay Stephenson]

Okay Cindy:

You probably have 2 lpsched running.

Do this:

lpshut (to kill the 'good' one)
ps -e | grep lpsched
If you find another lpsched pid, kill it like this:
kill PID
if that doesn't kill it, try kill -11 PID
finally try kill -9 PID but only after kill -11 has failed.

Next cd to /var/spool/lp
ls FIFO SCHEDLOCK
if either of these are found, rm them.
lpsched

Hopefully, that will fix you.

[Rita C Workman]

If your on 11.0 check and see if you have patch PHCO_22365. You may want to review the defect description on this...

Just a thought,
Rita

...and it does not require a reboot.

[Kevin Wright]

did you try kill on the PID? if not try it, first just do kill , then if that doesn't work, try kill -9 . If this doesn't work, you may have to reboot. you can check the pri by doing a ps -elf, if the pri is between 128 and 153, then you cannot send a signal to the process and may have to reboot.

[Cindy Wolford]

THANK YOU !!!!!!!

I did have a second lpsched running and by stopping the first, then killing the second, it worked!

THANK YOU !!!!!!!!!!!!!1
Cindy

[Bill Hassell]

Just a note on the design of lpsched. The first copy becomes a daemon to monitor all the print jobs. Each print job will start another lpsched which in turn starts the printer script. So it is quite normal to have two or more lpsched processes.

When lpshut is run, the daemon signals all the lpscheds to terminate. If an lpsched gets hung up, then two or more will exist until the child lpsched's are killed. The daemon lpsched has a parent PID of 1.